Overview and Methodology

Measuring the New Economy at the national level is not an easy task. The federal statistical system, which was founded largely on the notion of a stable economy with most of the output in agricultural and manufactured goods, still tends to focus on monetary measures related to managing the business cycles. But the New Economy is not stable, and boosting per capita income growth, rather than just managing the business cycle, needs to be the new focus of economic policy.2

PPI attempted to illustrate what is actually new about this so-called New Economy in The New Economy Index: Understanding America's Economic Transformation.3 In that report, the Institute used indicators gathered from disparate public and private data sources to track the structural transformation of the U.S. economy along four main lines: the industrial and occupational mix, globalization, entrepreneurial dynamism and competition, and the IT revolution. PPI's The State New Economy Index built on this work, applying key measures of the New Economy to state economies.4

But measuring the New Economy at the state and metropolitan level is even more difficult than it is at the national level, because many of the most useful data tend to be nationally, as opposed to regionally, oriented. Given that regional clusters of innovation play a more important role in the New Economy, this gap makes gaining a detailed examination of the New Economy all the more difficult.

Due to data limitations, there are a number of New Economy factors that should be included in this study but were not. For example, while data are available on high-tech industries, data are not available on the degree to which a metro area's older industries are using advanced technologies to produce and deliver goods and services. Similarly, while data measuring the educational attainment of an area's workforce are available, there are no data measuring the degree to which an area's companies are training their workers or reorganizing work to become high-performance organizations - both of which are important factors in determining how in sync an area's companies are with the New Economy. Data on exports are only available for manufacturing, not services, while data on foreign direct investment and other indicators of globalization (e.g., international telecommunications traffic) are not available. Similarly, accurate and consistent data on advanced metro area telecommunications infrastructures are limited, while data on industrial R&D are absent.

Moreover, not all indicators used in this report are perfect measures of New Economy characteristics. For example, the indicator of export orientation of manufacturing favors metro areas whose manufacturing sectors have become global (a basic New Economy trait), but metro areas like Richmond, Va., which exports a large amount of tobacco products, also get high marks based on their old-economy strengths. However, despite these limitations, a number of factors can still be measured that, we believe, collectively paint a robust picture for comparing metropolitan area economies.

For this study, we collected data on the 50 largest consolidated metropolitan statistical areas (CMSAs) in the United States as defined by the Federal Office of Management and Budget in 1999. This leads to two questions: Why look at metropolitan areas? And why use consolidated metropolitan areas instead of the smaller federal definition known as primary metropolitan areas?

The answer to the first question is easy. Over the last few years, metropolitan areas have increasingly been viewed as the fundamental building blocks of the global economy. As trade barriers have fallen under NAFTA and GATT, the nation-state has receded in importance as a "natural" unit of economic analysis. Although important from a political perspective, the 50 states and the District of Columbia have never been a particularly logical scale at which to analyze economic phenomena. The boundaries between the District of Columbia, Maryland, and Virginia are completely arbitrary in economic terms, while the residents of the northern and southern halves of California, Illinois, and New Jersey think of themselves as living in different economies (indeed, in different cultures).

The metropolitan area, on the other hand, is defined by the Census Bureau as a set of counties that exceed a certain threshold of commuting to the closest central city. A metropolitan area is, for the most part, a single labor market, with a one- to two-hour commute from edge to edge. Specialized job skills needed by specific industries, like pharmaceuticals in northern New Jersey or software in Seattle, are accessed easily within metropolitan areas, but are extremely difficult to trade between them. The New Economy is knowledge- and skill-driven, and metropolitan areas are the scale at which these specialized skills live.

Another conceptual origin of metropolitan areas lies not in the labor market, but in the relationships among firms in related industries. In the New Economy, such collaboration among spatially proximate firms (and other supporting institutions) has become increasingly important to economic innovation. These transactions are less frequent, so they can occur over greater distances. That is why it makes sense for the purpose of this study to use the larger of the two metropolitan area definitions: the CMSA, which includes the county "hinterlands" of two or more large central cities that are adjacent to each other. In New Economy terms, it seems clear that the primary metropolitan areas of San Francisco, Oakland, and San Jose should be combined. The same goes for Denver and Boulder, Dallas and Fort Worth, Chicago and Gary. (Note: In the pages that follow, we will sometimes designate a CMSA by its largest city only. CMSAs can be quite extensive and include a number of cities that are not named.5

In looking at the 50 most populated metro areas, the authors use 16 indicators divided into five categories that best capture what is new about the New Economy:

1) Knowledge jobs. Indicator measures jobs held by managers, professionals, and technicians; and the educational attainment of the workforce.

2) Globalization. Indicator measures the export orientation of manufacturing.

3) Economic dynamism and competition. Indicators in this category measure the number of fast-growing "gazelle" companies (companies with sales growth of 20 percent or more for four straight years); the rate of economic "churn" (which is a product of new business start-ups and existing business failures); and the number of initial public stock offerings (IPOs) by companies in each metro.

4) The transformation to a digital economy. Indicators measure the percentage of adults online; the number of ".com" domain-name registrations; the share of students using computers in schools; Internet backbone capacity; and number of providers of broadband telecommunications services.

5) Technological innovation capacity. Indicators measure the number of high-tech jobs; the number of science and engineering graduates from area colleges and universities; the number of patents issued; expenditures on research and development at colleges and universities; and venture capital investments.

In all cases, the report relies on the most recently published statistics available, but because of the delays in publishing statistics, particularly federal, the data may in some cases be several years old. In addition, in all cases data are reported to control for the size of the metropolitan area, using factors such as the number of workers or gross metropolitan product (gmp) as the denominator. For some indicators, data were missing or incomplete for a few metropolitan areas. In these cases, we used a number of different measures to estimate the scores and we indicate where this has been done by asterisks next to the indicator on the two large summary tables.

The overall New Economy scores were calculated as follows. In order to measure the magnitude of the differences between the metro areas, instead of just their rank from one to 50, raw scores are based on standard deviations from the mean. Thus, the raw score was first calculated (e.g., venture capital as a share of gross metropolitan product). Then, the mean score for each of the 50 metros was calculated and each score's deviation from the mean was calculated. Therefore, on most indicators, approximately half the metro areas have negative scores (below the 50-metro mean) and approximately half have positive scores. Using standard deviations accounts not just for the rank, but for the relative difference between scores, giving more weight, for example, to a metro that scores significantly above others, as compared to one that is only marginally above others.

In three of the five sub-index categories, and in the calculation of the overall New Economy scores, the indicators are weighted so that closely correlated ones (for example, patents, R&D spending, and high-tech workers) do not bias the results for the overall scores. (See Appendix.)

The overall scores are calculated by adding the metros' adjusted scores in each of the five sub-index categories. The sum of the individual indicator scores are equally adjusted (20 is added to every final metro score) to ensure that all are positive. These final scores are then divided by the sum of the highest score achieved by any metro in each category. Thus, each metro's final score is a percentage of the total score a metro would have achieved if it had finished first in every category.

The Progressive Policy Institute (PPI)
Technology, Innovation, and New Economy Project
600 Pennsylvania Ave., S.E., Suite 400, Washington DC 20003
Phone: (202) 547-0001
www.ppionline.org

Website design by OnlineWorkshop.

Overview & Methodology